Knee osteoarthritis (OA) is the most common cause of chronic disability among the elderly in the United States. In a recent study of community-based subjects with knee OA, disability was more closely associated with quadriceps muscle weakness than with radiographic severity of disease or severity of joint pain. Although it has been widely considered that quadriceps weakness in subjects with knee is due to disuse atrophy secondary to joint pain, subjects with knee OA have been described who do not have knee pain but in whom weakness appears to result from an inability to fully contract the quadriceps. This phenomenon, termed arthrogenous muscle inhibition (AMI), has been attributed to neuronal reflex activity in which altered afferent input from the OA joint results in diminished efferent motor drive to periarticular muscles. It is a specific aim of this project to conduct a cross-sectional correlational study in which the level of AMI in both knees of patients with radiographic evidence of knee OA, encompassing a broad range of severity, will be determined by a twitch superimposition technique and related to: quadriceps strength, lean tissue (i.e., muscle) mass in the thigh (estimated by dual energy x- ray absorptiometry), duration of disease, radiographic severity of OA, level of joint pain and function [assessed by the Western Ontario and McMasters Osteoarthritis Index (WOMAC)], lower extremity performance tests (50' walk time, timed stands test), physical findings (knee tenderness, effusion), depression, level of social support, and use of nonsteroidal anti-inflammatory drugs, analgesics and psychotropic agents. Cross-sectional data from 60 patients will be supplemented by data from 60 community-based subjects with radiographic evidence of knee OA whom we have found to have quadriceps weakness, but who do not have significant functional impairment, joint pain or NSAID intake. Our second specific aim is to conduct a longitudinal study to test the hypothesis that the baseline level of AMI in individuals with knee OA predicts the severity of subsequent lower extremity functional impairment, knee pain and reduction in quadriceps strength and mass (i.e., atrophy). Accordingly, 6 mos. and 12 mos. after the initial examination we will re-assess those patients who, in the baseline analysis, exhibit only mild-moderate functional impairment. In addition, we will attempt to re-evaluate all available subjects from the community cohort. This approach should permit us to define changes in AMI over time in individuals with knee OA and to ascertain the importance of AMI as a risk factor for disability, knee pain, quadriceps weakness and quadriceps atrophy. If AMI is established as a risk factor for the above consequences of knee OA, the results will provide the basis for future interventional studies aimed at preventing or reversing AMI through pharmacological or physical measures.